Automated identification of magnetic responses from porphyry systems

Abstract

Porphyry-style mineralisation often appears as near circular anomalies within magnetic data. This article presents an automatic grid analysis system to detect such responses. Our approach follows three steps: (1) Find circular features using the radial symmetry transform; (2) Validate the detected features by the presence of high magnetic contrast at the feature location; and (3) Highlight the alteration zone boundary using deformable splines. The outcome of this system is two-fold. First, provide estimates of the location of potential porphyry systems, and second, accentuate the appearance of potential exploration targets to aid manual data inspection. Experiments were conducted on survey data from Reko Diq, Pakistan, a region known to contain numerous occurrences of porphyry-style mineralisation. The prediction results of our system closely matched the location of the known deposits in this region thus rendering confidence in the effectiveness of our approach. It is suggested that this system be used as an initial screening tool for large datasets, therefore reducing the time and cost imposed by manual data inspection in the exploration targeting process.